Dissertations / Theses on the topic 'Bayesian belief network'
To see the other types of publications on this topic, follow the link: Bayesian belief network.
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 dissertations / theses for your research on the topic 'Bayesian belief network.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
1
Pershad, Rinku. "A Bayesian belief network for corporate credit risk assessment." Thesis, National Library of Canada = Bibliothèque nationale du Canada, 2000. http://www.collectionscanada.ca/obj/s4/f2/dsk1/tape4/PQDD_0022/MQ50360.pdf.
Full text
2
Sahely, Brian S. G. E. "Development of a Bayesian belief network for anaerobic wastewater treatment." Thesis, National Library of Canada = Bibliothèque nationale du Canada, 2000. http://www.collectionscanada.ca/obj/s4/f2/dsk1/tape4/PQDD_0027/MQ50490.pdf.
Full text
3
Ejaz, Azad. "Using a Bayesian Belief Network for Going-Concern Risk Evaluation." NSUWorks, 2005. http://nsuworks.nova.edu/gscis_etd/500.
Full textAbstract:
An auditor's verdict on client's financial health is delivered in the form of a going concern (GC) opinion. Although an auditor is not required to predict the financial future of a client, stakeholders take the GC opinion as a guideline on a company's financial health. The GC opinion has been a subject of much debate in the financial literature, as it is one of the most widely read parts of an audit report. Researchers and academicians believe that auditors have made costly mistakes in rendering GC opinions. Several factors have been identified as the root causes for these mistakes, including growing business complexities, insufficient auditor training, internal and external pressures, personal biases, economic considerations, and fear of litigation. To overcome these difficulties, researchers have been trying to devise effective audit tools to help auditors form accurate GC opinions on clients ' financial future. Introduction of ratio-based bankruptcy models using a variety of statistical techniques are attempts in the right direction. The results of such efforts, though not perfect, are encouraging.
This study examined several popular ratio-based statistical models and their weaknesses and limitations. The author suggests a new model based on the robust Bayesian Belief Network (BBN) technique. Based on sound Bayesian theory, this model provides remedies against the reported deficiencies of the ratio-based techniques. The proposed system, instead of comparing a company's financial ratios with the industrywide ratios, measures the internal financial changes within a company during a particular year and uses the changing financial pattern to predict the financial viability of the company. Unlike other popular models, the proposed model takes various qualitative factors into consideration before delivering the GC verdict. The proposed system is verified and validated by comparing its results with the industry de facto Z-score model.
4
Leerojanaprapa, Kanogkan. "A Bayesian belief network modelling process for systemic supply chain risk." Thesis, University of Strathclyde, 2014. http://oleg.lib.strath.ac.uk:80/R/?func=dbin-jump-full&object_id=23564.
Full textAbstract:
To effectively manage risk in supply chains, it is important to understand the interrelationships between risk events that might affect the flow of material, products and information within the chain. Typical supply chain risk management tends to treat events as if they are independent and so fail to capture the systemic nature of supply chain risks. This thesis addresses this shortcoming by developing a quantitative modelling process to support systemic supply chain risk analysis. Bayesian Belief Network (BBN) models are able to capture both the aleatory and epistemic uncertainties associated with supply chains and to represent probabilistic dependency relationships. A visual modelling process, grounded in the theory of BBN and the decision context of supply chain risk management, is developed to capture the knowledge and probability judgements of relevant stakeholders. An experiment has been conducted to evaluate alternative approaches to structuring a BBN model for supply risk. It is found that building causal maps provides a good basis for translating stakeholder cause-effect knowledge about the supply chain risks into a formal graphical probability model, which underpins the BBN. The modelling process has been evaluated through a longitudinal case for the hospital medicine supply of NHS Greater Glasgow & Clyde. A BBN model has been developed in collaboration with relevant stakeholders who have expertise in all or part of the medicine supply chain. The perceptions of these stakeholders about the modelling process and results generated have been formally gathered and analysed. The BBN model of the medicine supply chain has provided insight into risks not captured by conventional risk management methods and supported deeper understanding of risk through exploration of modelling scenarios. Analysis of stakeholder evaluation of the modelling process provided valuable insights into the operationalization of BBN modelling for supply risk and has informed the final modelling process developed through this research.
5
Ang, Kwang Chien. "Applying Bayesian belief networks in Sun Tzu's Art of war." Thesis, Monterey, California. Naval Postgraduate School, 2004. http://hdl.handle.net/10945/1323.
Full textAbstract:
Approved for public release; distribution in unlimited.The principles of Sun Tzu's Art of War have been widely used by business executives and military officers with much success in the realm of competition and conflict. However, when conflict situations arise in a highly stressful environment coupled with the pressure of time, decision makers may not be able to consider all the key concepts when forming their decisions or strategies. Therefore, a structured reasoning approach may be used to apply Sun Tzu's principles correctly and fully. Sun Tzu's principles are believed to be able to be modeled mathematically; hence, a Bayesian Network model (a form of mathematical tool using probability theory) is used to capture Sun Tzu's principles and provide the structured reasoning approach. Scholars have identified incompleteness in Sun Tzu's appreciation of information in war and his application of secret agents. This incompleteness resulted in circular reasoning when both sides of the conflict apply his principles. This circular reasoning can be resolved through the use of advanced probability theory. A Bayesian Network Model however, not only provides a structured reasoning approach, but more importantly, it can also resolve the circular reasoning problem that has been identified.
Captain, Singapore Army
6
Nunoo, Samuel. "Bayesian Belief network approach to slope management in British Columbia open pits." Thesis, University of British Columbia, 2016. http://hdl.handle.net/2429/57946.
Full textAbstract:
The stability of rock slopes is a major safety issue in open pit mining. It is important for rock engineers and mine operators to be knowledgeable about their pit wall behaviour, and, more specifically, to recognize appropriate conditions that trigger the need to issue warnings or stop work orders. With the current increase in the number of open pit mines in British Columbia and the deepening of existing pits, there is a need for rational, scientifically based decisions in response to measured pit wall performance. The main objective of this research was to develop and establish a Bayesian Belief Network (BBN) model and outline appropriate operational responses to manage slopes in large open pit porphyry mines. The BBN model can be tailored to specific geotechnical conditions and pit wall configurations. The research integrated available geotechnical engineering data and knowledge, including expert knowledge, ground water conditions, slope geometry, mining activity (blast damage), and consequences of failure, into one platform that can establish appropriate operational responses. A range of pre-defined actions ranging from normal pit operations to orders to stop work and evacuate the pit were defined in this research as operational responses or pit management decisions. These operational responses were linked in the BBN model to predicted states of pit wall movement and estimates of the consequences of these movements. A new relationship was proposed to estimate the travel distance from a wide range of pit slope failure debris volumes. The relationship accounts for a potential rockslide transforming into a rock avalanche. The BBN model was used to retroactively predict the appropriate operational response at four mines to using data from past slope instabilities. The results indicate that equipment damage as well as production losses could have been minimized or prevented had the BBN model been used by the mine operators at the time of each slope instability. The methodology described in the thesis provides the foundation for an innovative tool for the selection of appropriate operational responses linked to measured slope velocity, potential rockslide debris volume, and potential travel distance of the debris.Applied Science, Faculty of
Engineering, School of (Okanagan)
Graduate
7
Lee, Keen Sing 1972. "Quantifying the Main Battle Tank's architectural trade space using Bayesian Belief Network." Thesis, Massachusetts Institute of Technology, 2004. http://hdl.handle.net/1721.1/34733.
Full textAbstract:
Thesis (S.M.)--Massachusetts Institute of Technology, System Design & Management Program, 2004.Includes bibliographical references (p. 239-240).
The design and development of a Main Battle Tank can be characterized as a technically challenging and organizationally complex project. These projects are driven not only by the essential engineering and logistic tasks; as the frequency of technological innovation increases system architects are motivated to apply an effective method to assess the risks and benefits of adopting technological alternatives. This thesis applies Bayesian Belief Network as a quantitative modeling and metrics calculation framework in establishing the preference order of possible architectural choices during the development of a Main Battle Tank. A framework of metrics was developed for the architect to communicate objectively with stakeholders and respond to challenges raised. These inputs were then encoded as variables in a global Bayesian Belief Network. Using a change propagation algorithm any changes in the probabilities of individual variables would trigger changes throughout the entire network and can be used as informing messages to the stakeholders to reflect the consequences of these changes. Two Bayesian Belief Networks were developed and tested to understand the effectiveness and sensitivities to the variables. The successful development of the Bayesian Belief Network offers technical and organizational benefits to the system architect. From the technical viewpoint, the model benefits include performing system tradeoff studies, iterating the design to incorporate feedback quickly, analyzing the sensitivity and impact of each design change to the overall system, and identifying critical areas to allocate resources. From an organizational process perspective, it enables speedier knowledge transfer in the project, and enables the engineers
(cont.) to be knowledgeable about how their localized change could affect other sub-systems.
by Keen Sing Lee.
S.M.
8
Gilson, Robert. "Minimizing input acquisition costs in a Bayesian belief network-based expert system /." Thesis, Connect to this title online; UW restricted, 1997. http://hdl.handle.net/1773/8763.
Full text
9
Kim, Dohyoung 1970. "Bayesian Belief Network (BBN)-based advisory system development for steam generator replacement project management." Thesis, Massachusetts Institute of Technology, 2002. http://hdl.handle.net/1721.1/30011.
Full textAbstract:
Thesis (Sc. D.)--Massachusetts Institute of Technology, Dept. of Nuclear Engineering, 2002.Includes bibliographical references (leaves 192-194).
The growing need for improved project management technique points to the usefulness of a knowledge-base advisory system to help project managers understand current and future project status and optimize decisions based upon the project performances. The work here demonstrates the framework of an advisory system with improved ability in project management. Based upon the literature survey and discussion with relevant experts, the Bayesian Belief Network (BBN) approach was selected to model the steam generator replacement proj ect management problem, where the situation holds inherently large uncertainty and complexities, since it has a superior ability to treat complexities, uncertainty management, systematic decision making, inference mechanism, knowledge representation and model modification for newly acquired knowledge. Two modes of advisory system have been constructed. As the first mode, the predictive mode has been developed, which can predict future project performance state probability distributions, assuming no intervening management action. The second mode is the advisory mode, which can identify the optimal action among alternatives based upon the expected net benefit values that are incorporating two important components: 1) expected immediate net benefits at post-action time, and 2) the expected long term benefit (or penalty) at scheduled project completion time. During the work, new indices for important variables have been newly developed for effective and efficient project status monitoring. With application of developed indices to the advisory system, the long term benefit (or penalty) found to be the most important factor in determining the optimal action by the project management during the decision
(cont.) making process and was confirmed by the domain experts. As a result, the effort has been focused on incorporating the long term benefit (or penalty) concept in order to provide more reliable and accurate advice to the project managers. In addition, in order to facilitate the communication between the BBN models and the users, an interface program has been developed using the Visual Basic language.
by Dohyoung Kim.
Sc.D.
10
REN, Qing. "Applying Bayesian Belief Network To Understand Public Perception On Green Stormwater Infrastructures In Vermont." ScholarWorks @ UVM, 2018. https://scholarworks.uvm.edu/graddis/835.
Full textAbstract:
Decisions of adopting best management practices made on residential properties play an important role in reduction of nutrient loading from non-point sources into Lake Champlain and other waterbodies in Vermont. In this study, we use Bayesian belief network (BBN) to analyze a 2015 survey dataset about adoption of six types of green infrastructures (GSIs) in Vermont’s residential areas. Learning BBNs from physical probabilities of the variables provides a visually explicit approach to reveal the message delivered by the dataset. Using both unsupervised and supervised machine learning algorithms, we are able to generate networks that connect the variables of interest and conduct inference to look into the probabilistic associations between the variables. Unsupervised learning reveals the underlying structures of the dataset without presumptions. Supervised learning provides insights for how each factor (e.g. demographics, risk perception, and attribution of responsibilities) influence individuals’ pro-environmental behaviors. We also compare the effectiveness of BBN approach and logistic regression in predicting the pro-environmental behaviors (adoption of GSIs).
The results show that influencing factors for current adoption vary by different types of GSI. Risk perception of stormwater issues are associated with adoption of GSIs. Runoff issues are more likely to be considered as the governments’ (town, state, and federal agencies) responsibility, whereas lawn erosion is more likely to be considered as the residents’ own responsibility. When using the same set of variables to predict pro-environmental behaviors (adoption of GSI), BBN approach produces more accurate prediction compared to logistic regression.
11
Franco, Chiara. "Modelling the dynamics of CaCO3 budgets in changing environments using a Bayesian Belief Network approach." Thesis, University of Essex, 2014. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.654563.
Full textAbstract:
Multiple stressors on reefs are increasing the need for remedial actions to buffer anthropogenic pressure and reduce coral reef deterioration. In order to promote reef framework endurance, it is critical to identify and track down multiple stressor sources. To date, spatial and temporal variations of reef framework carbonate production and erosion have been estimated using carbonate budget assessments; however, these are limited in determining the extent to which various stressors are responsible for altering the budgetary state. This study has developed a Bayesian Network model (CARBNET) to identify and evaluate the extent to which anthropogenic and climatic disturbances affect coral reef budgetary state. The main adavantage of using this type of model for management purposes are related to its ability to adapt to changes and to quantify and incorporate uncertainty. In addition, it provides the opportunity to identify key gaps in the knowledge to inform future research priorities. Multi-scale scenario-based analyses, conducted for the Wakatobi (South-east Sulawesi, Indonesia) and Grenada (Caribbean) reefs, quantified the effects of multiple stressors on the reefal components, providing information on the actual state and possible future state of the framework. Reefs with high branching coral cover were likely to be found in a positive budgetary state, whilst low coral cover and reduced topographic complexity were associated with low carbonate production or negative budgetary state. In clear water settings, degraded reefs, characterised by high turbidity, sedimentation and nutrient concentrations, were likely to be found in a low carbonate production or erosional state. Conversely, high carbonate production was characteristic of reef environments with low turbidity, sedimentation and nutrient concentrations. At regional level, CARBNET predicted that reefs will accrete at a different pace; in Grenada reduced gross production and sustained erosion maintained the budget close to the equilibrium, whilst Wakatobi reefs were defined by positive budgetary states. At local level, reefs at shallow depths were likely to be associated with erosion or low positive net production in both regions, although in Indonesia high carbonate production offsets erosion at all sites. Anthropogenic and climatic disturbances acted synergistically in decreasing carbonate production, and degraded reefs with < 10% hard coral were predicted to be in an erosional state. This result suggests that degraded systems have lowered their tipping point to a net erosion shift. The benthic community was affected by sedimentation and elevated nutrients and changes in the key drivers of carbonate production resulted in reduced net carbonate production. External bioeroder densities were restrained by degradation of the nursery ground, whilst internal bioerosion increased in nutrient enriched waters. Overall, CARBNET is reliable in groundtruthing empirical data and is therefore a valuable addition to the reef management toolbox.
12
Oteniya, Lloyd. "Bayesian belief networks for dementia diagnosis and other applications : a comparison of hand-crafting and construction using a novel data driven technique." Thesis, University of Stirling, 2008. http://hdl.handle.net/1893/497.
Full textAbstract:
The Bayesian network (BN) formalism is a powerful representation for encoding domains characterised by uncertainty. However, before it can be used it must first be constructed, which is a major challenge for any real-life problem. There are two broad approaches, namely the hand-crafted approach, which relies on a human expert, and the data-driven approach, which relies on data. The former approach is useful, however issues such as human bias can introduce errors into the model. We have conducted a literature review of the expert-driven approach, and we have cherry-picked a number of common methods, and engineered a framework to assist non-BN experts with expert-driven construction of BNs. The latter construction approach uses algorithms to construct the model from a data set. However, construction from data is provably NP-hard. To solve this problem, approximate, heuristic algorithms have been proposed; in particular, algorithms that assume an order between the nodes, therefore reducing the search space. However, traditionally, this approach relies on an expert providing the order among the variables --- an expert may not always be available, or may be unable to provide the order. Nevertheless, if a good order is available, these order-based algorithms have demonstrated good performance. More recent approaches attempt to ''learn'' a good order then use the order-based algorithm to discover the structure. To eliminate the need for order information during construction, we propose a search in the entire space of Bayesian network structures --- we present a novel approach for carrying out this task, and we demonstrate its performance against existing algorithms that search in the entire space and the space of orders. Finally, we employ the hand-crafting framework to construct models for the task of diagnosis in a ''real-life'' medical domain, dementia diagnosis. We collect real dementia data from clinical practice, and we apply the data-driven algorithms developed to assess the concordance between the reference models developed by hand and the models derived from real clinical data.
13
Tang, Antony Shui Sum, and n/a. "A rationale-based model for architecture design reasoning." Swinburne University of Technology, 2007. http://adt.lib.swin.edu.au./public/adt-VSWT20070319.100952.
Full textAbstract:
Large systems often have a long life-span and their system and software architecture design comprise many intricately related elements. The verification and maintenance of these architecture designs require an understanding of how and why the system are constructed. Design rationale is the reasoning behind a design and it provides an explanation of the design. However, the reasoning is often undocumented or unstructured in practice. This causes difficulties in the understanding of the original design, and makes it hard to detect inconsistencies, omissions and conflicts without any explanations to the intricacies of the design. Research into design rationale in the past has focused on argumentation-based design deliberations. Argumentation-based design rationale models provide an explicit representation of design rationale. However, these methods are ineffective in communicating design reasoning in practice because they do not support tracing to design elements and requirements in an effective manner.
In this thesis, we firstly report a survey of practising architects to understand their
perception of the value of design rationale and how they use and document this knowledge.
From the survey, we have discovered that practitioners recognize the importance of documenting design rationale and frequently use them to reason about their design choices. However, they have indicated certain barriers to the use and documentation of design rationale. The results have indicated that there is no systematic approach to using and capturing design rationale in current architecture design practice. Using these findings, we address the issues of representing and applying architecture design rationale.
We have constructed a rationale-based architecture model to represent design rationale,
design objects and their relationships, which we call Architecture Rationale and
Element Linkage (AREL). AREL captures both qualitative and quantitative rationale for
architecture design. Quantitative rationale uses costs, benefits and risks to justify architecture
decisions. Qualitative rationale documents the issues, arguments, alternatives and
tradeoffs of a design decision. With the quantitative and qualitative rationale, the AREL
model provides reasoning support to explain why architecture elements exist and what
assumptions and constraints they depend on. Using a causal relationship in the AREL
model, architecture decisions and architecture elements are linked together to explain the reasoning of the architecture design. Architecture Rationalisation Method (ARM) is a
methodology that makes use of AREL to facilitate architecture design. ARM uses cost,
benefit and risk as fundamental elements to rank and compare alternative solutions in the decision making process.
Using the AREL model, we have proposed traceability and probabilistic techniques
based on Bayesian Belief Networks (BBN) to support architecture understanding and
maintenance. These techniques can help to carry out change impact analysis and rootcause analysis. The traceability techniques comprise of forward, backward and evolution tracings. Architects can trace the architecture design to discover the change impacts by analysing the qualitative reasons and the relationships in the architecture design. We have integrated BBN to AREL to provide an additional method where probability is used to evaluate and reason about the change impacts in the architecture design. This integration provides quantifiable support to AREL to perform predictive, diagnostic and combined reasoning.
In order to align closely with industry practices, we have chosen to represent the
rationale-based architecture model in UML. In a case study, the AREL model is applied
retrospectively to a real-life bank payment systems to demonstrate its features and applications.
Practising architects who are experts in the electronic payment system domain
have been invited to evaluate the case study. They have found that AREL is useful in
helping them understand the system architecture when they compared AREL with traditional design specifications. They have commented that AREL can be useful to support the verification and maintenance of the architecture because architects do not need to reconstruct or second-guess the design reasoning.
We have implemented an AREL tool-set that is comprised of commercially available
and custom-developed programs. It enables the capture of architecture design and its
design rationale using a commercially available UML tool. It checks the well-formedness
of an AREL model. It integrates a commercially available BBN tool to reason about the
architecture design and to estimate its change impacts.
14
Spenkuch, Thomas. "A Bayesian belief network approach for modelling tactical decision-making in a multiple yacht race simulator." Thesis, University of Southampton, 2014. https://eprints.soton.ac.uk/366587/.
Full textAbstract:
The importance of human factors has to be taken into account when determining a yacht’s performance over a course. The crew’s capabilities of technical skills, athletic performance, and his/her ability of making rational decisions under time pressure and in light of uncertainty of the future wind regime are important aspects that will determine the overall performance of a yacht-crew system. This thesis highlights the performance of such a yacht-crew system with a focus on the decision-making process of sailors. Aspects of human behaviour in sport and the decision-making process are explained considering the level of expertise and possible approaches of how to model them are shown. An artificial intelligence AI -system is developed that is capable of simulating the decision-making process of different sailing behaviours/styles as well as different expertise levels of sailors within a dynamically changing yacht racing environment. The constraints of the multiple fleet racing simulator Robo-Race (Scarponi 2008) were determined using a series of tests with real sailors identified three important constrains: (1) the predictable behaviour of the AI-yachts, (2) the predictable and unrealistic weather model and (3) the simple model describing the effects of yacht interaction. These restrictions and constraints that limited the real and AI-sailors natural sailing behaviour have been successfully removed in the updated version of Robo-Race. The new developed decision-making engine based on Decision Field Theory that uses Bayesian Belief Networks as the perceptual processor showed a clear superiority over the old rule-based decision-making engine. Extensive simulations demonstrate the feasibility of modelling various decision-making processes and therefore different behaviours and expertise levels of sailors. A good comparison was found with that obtained between the Robo-Race results and the Olympic fleet racing events.
15
Shabarchin, Oleg. "Induced seismicity and corrosion vulnerability assessment of oil and gas pipelines using a Bayesian belief network model." Thesis, University of British Columbia, 2016. http://hdl.handle.net/2429/57569.
Full textAbstract:
A substantial amount of oil and gas products are transported and distributed via pipelines, which can stretch for thousands of kilometers. Because of the adverse environmental impact and significant financial losses, the integrity of these pipelines is essential. British Columbia Oil and Gas Commission (BCOGC) has indicated metal loss due to corrosion as one of the primary causes of pipeline failures. Therefore, it is important to identify pipelines subjected to severe corrosion in order to improve corrosion mitigation and pipeline maintenance strategies, thus minimizing the likelihood of failure. To accomplish this task, this thesis presents a Bayesian belief network (BBN)-based probabilistic corrosion hazard assessment approach for oil and gas pipelines. A cause-effect BBN model has been developed by considering various types of information, such as analytical corrosion models, expert knowledge and published literature. Multiple corrosion models and failure pressure models have been incorporated into a single flexible network in order to estimate corrosion defects and the associated probability of failure. Besides corrosion hazard, BCOGC has reported multiple cases of anthropogenic seismicity, which also may compromise the pipeline integrity. To this end, this thesis explores the potential impact of induced seismicity on the oil and gas pipeline infrastructure. Spatial clustering analysis is used for earthquakes, previously registered in the region, to delineate areas, which are particularly prone to the induced seismicity. The state of the art ground motion prediction equation for induced seismicity is applied in a Monte Carlo simulation to obtain a stochastic field of the seismic intensity. Based on the established seismic fragility formulations for pipelines and mechanical characteristics as well as corrosion conditions, spatial and probabilistic distributions of the repair rate and probability of failure have been obtained and visualized with the aid of the Geographic Information System. The proposed model can help to identify vulnerable pipeline sections and rank them accordingly to enhance the informed decision making process. To demonstrate the application of the proposed approach, two case studies for the Northeastern British Columbia oil and gas pipeline infrastructure are presented.Applied Science, Faculty of
Engineering, School of (Okanagan)
Graduate
16
Morgenroth, Josephine S. "Elastic stress modelling and prediction of ground class using a Bayesian Belief Network at the Kemano tunnels." Thesis, University of British Columbia, 2016. http://hdl.handle.net/2429/58971.
Full textAbstract:
The Kemano hydroelectric facility was constructed in the 1950s to supply power to the
aluminum smelter in Kitimat, on the west coast of British Columbia. The Kemano project
includes a 16 km long water conveyance tunnel that set world record advance rates in the
1950s, and 8 km of a partially completed tunnel.
A risk management strategy was developed in the late 1980s in case of collapse of the first water conveyance tunnel, and by 1990 the excavation of a second tunnel parallel to the first had begun. Work halted in 1991 due to environmental litigation and change in political climate. In 2011 the owner of the Kitimat smelter and the Kemano hydroelectric facility announced plans to continue work on the tunnel that was left unfinished. This thesis is a collaboration with Hatch Ltd., a consultant to the owner, to determine the ground conditions and support requirements that should be anticipated in completing the backup tunnel. Three dimensional finite element elastic stress modelling was completed in order to determine the in-situ stress conditions as well as the boundary stresses around the tunnel. The modelling results were used to estimate where stress-induced problem areas should be expected, for example at chainages 10+700 to 12+700 in the backup tunnel. The results of the stress modelling were incorporated into a Bayesian Belief Network that was developed for the Kemano tunnels. It was built using widely accepted empirical relationships in rock mechanics, expert judgement and conditional relationships between inputs. This network predicts the ground class at a user-defined chainage, based on a database that was developed from project literature. The user is also able to input new data
as it becomes available, for example during the tunnel advance. The predictions from the network align with what can be seen in the excavated portion of the backup tunnel, for example accurately predicting the need for steel sets at chainage 8+510. The predicted ground class was plotted as a function of chainage, and may be used as a comparison to the support requirements that have been determined thus far.Applied Science, Faculty of
Engineering, School of (Okanagan)
Graduate
17
Dumas, Jeremiah Percy. "A spatial decision support system utilizing data from the Gap Analysis Program and a Bayesian Belief Network." Master's thesis, Mississippi State : Mississippi State University, 2005. http://library.msstate.edu/etd/show.asp?etd=etd-07072005-104946.
Full text
18
Li, Mingmei. "Application of the Bayesian belief network model to evaluate variances in a clinical caremap: Radical prostatectomy case study." Thesis, University of Ottawa (Canada), 2004. http://hdl.handle.net/10393/26693.
Full textAbstract:
A clinical caremap is a cost-effective tool for clinical process improvement that has been accepted in hospitals and various healthcare organizations in many countries. However, compared to the literature describing the initial development of the clinical caremaps, the evaluation of the impact of the variances in the caremap pathway on the patient's expected outcomes and the patient's length of stay (LOS) remains relatively less analyzed. In this research, we deal with the issue of variances in the clinical caremap by building a Bayesian belief network named BBN_RPC to model the radical prostatectomy caremap. The BBN_RPC model provides insight into probabilistic dependencies that exist among the activities (variables) in the caremap. We then use the BBN_RPC model to analyze possible variances and to make inferences. The results show that most of the activities in the caremap are related with each other and to some extent linked with the patient's length of stay (LOS), whereas different activities have different weights on the LOS. Using radical prostatectomy patients' data from a retrospective chart study conducted at the Ottawa Civic Hospital, we have applied the BBN_RPC model to predict a patient's future conditions and the LOS, based on the current observations. Predictive accuracy of the BBN_RPC model was evaluated by cross validation tests, which showed the accuracy for predicting the patient's LOS, given the patient's observations during the first two post-op days, is at approximately 94% level.
19
Turschwell, Mischa Peter. "Thermal and Habitat Characteristics of a Headwater Fish Species: Predicting Population Success Under Climate Change." Thesis, Griffith University, 2017. http://hdl.handle.net/10072/367629.
Full textAbstract:
Streams are becoming increasingly fragmented by anthropogenic impacts related to altered thermal and hydrological regimes, dispersal barriers, land-use, introductions of non-native fish, habitat degradation, and altered disturbance regimes. Furthermore, relatively little is known about the ecology and spatio-temporal dynamics of a large proportion of freshwater fish populations and assemblages, making their conservation challenging. The general goal of my research was to improve ecological understanding of the drivers of fish distribution in upland streams, using the threatened Northern river blackfish (Gadopsis marmoratus - hereinafter blackfish) in the upper Condamine River and Spring Creek tributaries in Queensland, as a case study. While blackfish are still widespread throughout the Murray Darling Basin, anecdotal evidence suggests that historically, they were widely distributed throughout the entire upper catchment of the
Condamine River, extending into the lowlands. In addition, results from contemporary research suggests that these fish may now be restricted to headwaters and tributaries. The aim of my research was to identify the variables that influence blackfish distribution in this upland system. More specifically, I tested the hypothesis that these fish are thermally restricted to their current habitat range. I asked whether there are different processes governing their occurrence versus abundance, and whether these vary between life-stages. In addition to examining static population structure I also examined the environmental determinants of juvenile recruitment in this system, and examined the spatial scales at which these take place.Thesis (PhD Doctorate)
Doctor of Philosophy (PhD)
Griffith School of Environment
Science, Environment, Engineering and Technology
Full Text
20
Kim, Sojung. "Dynamic Learning and Human Interactions under the Extended Belief-Desire-Intention Framework for Transportation Systems." Diss., The University of Arizona, 2015. http://hdl.handle.net/10150/578837.
Full textAbstract:
In recent years, multi-agent traffic simulation has been widely used to accurately evaluate the performance of a road network considering individual and dynamic movements of vehicles under a virtual roadway environment. Given initial traffic demands and road conditions, the simulation is executed with multiple iterations and provides users with converged roadway conditions for the performance evaluation. For an accurate traffic simulation model, the driver's learning behavior is one of the major components to be concerned, as it affects road conditions (e.g., traffic flows) at each iteration as well as performance (e.g., accuracy and computational efficiency) of the traffic simulation. The goal of this study is to propose a realistic learning behavior model of drivers concerning their uncertain perception and interactions with other drivers. The proposed learning model is based on the Extended Belief-Desire-Intention (E-BDI) framework and two major decisions arising in the field of transportation (i.e., route planning and decision-making at an intersection). More specifically, the learning behavior is modeled via a dynamic evolution of a Bayesian network (BN) structure. The proposed dynamic learning approach considers three underlying assumptions: 1) the limited memory of a driver, 2) learning with incomplete observations on the road conditions, and 3) non-stationary road conditions. Thus, the dynamic learning approach allows driver agents to understand real-time road conditions and estimate future road conditions based on their past knowledge. In addition, interaction behaviors are also incorporated in the E-BDI framework to address influences of interactions on the driver's learning behavior. In this dissertation work, five major human interactions adopted from a social science literature are considered: 1) accommodation, 2) collaboration, 3) compromise, 4) avoidance, and 5) competition. The first three interaction types help to mimic information exchange behaviors between drivers (e.g., finding a route using a navigation system) while the last two interaction types are relevant with behaviors involving non-information exchange behaviors (e.g., finding a route based on a driver's own experiences). To calibrate the proposed learning behavior model and evaluate its performance in terms of inference accuracy and computational efficiency, drivers' decision data at intersections are collected via a human-in-the-loop experiment involving a driving simulator. Moreover, the proposed model is used to test and demonstrate the impact of five interactions on drivers' learning behavior under an en route planning scenario with real traffic data of Albany, New York, and Phoenix, Arizona. In this dissertation work, two major traffic simulation platforms, AnyLogic® and DynusT®, are used for the demonstration purposes. The experimental results reveal that the proposed model is effective in modeling realistic learning behaviors of drivers in conduction with interactions with other drivers.
21
Jönsson, Jack. "Belief-aided Robust Control for Remote Electrical Tilt Optimization." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-301028.
Full textAbstract:
Remote Electrical Tilt (RET) is a method for configuring antenna downtilt in base stations to optimize mobile network performance. Reinforcement Learning (RL) is an approach to automating the process by letting an agent learn an optimal control strategy and adapt to the dynamic environment. Applying RL in real world comes with challenges, for the RET problem there are performance requirements and partial observability of the system through exogenous factors inducing noise in observations. This thesis proposes a solution method through modeling the problem by a Partially Observable Markov Decision Process (POMDP). The set of hidden states are modeled as a high- level representation of situations requiring one of the possible actions uptilt, downtilt, no change. From this model, a Bayesian Neural Network (BNN) is trained to predict an observation model, relating observed Key Performance Indicators (KPIs) to the hidden states. The observation model is used for estimating belief state probabilities of each hidden state, from which decision of control action is made through a restrictive threshold policy. Experiments comparing the method to a baseline Deep Q- network (DQN) agent shows the method able to reach the same average performance increase as the baseline while outperforming the baseline in two metrics important for robust and safe control behaviour, the worst- case minimum reward increase and the average reward increase per number of tilt actions.Fjärrstyrning av Elektrisk Lutning (FEL) är en metod för att reglera lutningen av antenner i basstationer för att optimera presentandan i ett mobilnätverk. Förstärkande Inlärning (FI) används som metod för att automatisera processen genom att låta en agent lära sig en optimal strategi för reglering och anpassa sig till den dynamiska miljön. Att tillämpa FI i ett verkligt scenario innebär utmaningar, för FEL specifikt finns det krav på en viss nivå av prestanda samt endast en delvis observerbarhet av systemet på grund av externa faktorer som orsakar brus i observationerna. I detta arbete föreslås en metod för att hantera detta genom att modellera problemet som en Delvis Observerbar Markovprocess (DOM). De dolda tillstånden modelleras för att representera situationer där var och en av de möjliga aktionerna behövs, det vill säga att luta antennen upp, ner eller inte ändra på lutningen. Utifrån denna modellering så tränas ett Bayesiskt Neuralt Nätverk (BNN) för att estimera en observationsmodel som kopplar observerade nyckeltal till de dolda tillstånden. Denna observationsmodel används för att estimera sannolikheten att vardera dolt tillstånd är det rätta. Utifrån dessa sannolikheter så görs valet av aktion genom ett tröskelvärde på sannolikheterna. Genom experiment som jämför metoden med en standardimplementering av en agent baserad på ett Djupt Qnätverk (DQN) visas att metoden har samma prestation när det kommer till en medelnivå på prestandaökning i nätverket. Metoden överträffar dock standardmetoden i två andra mätvärden som är viktiga ur aspekten säker och robust reglering, minimumvärdet på prestandaökningen samt medelökningen av prestandan per antal up- och nerlutningar som används.
22
Renken, Hendrik. "Macroalgal dynamics on Caribbean coral forereefs." Thesis, University of Exeter, 2008. http://hdl.handle.net/10036/41253.
Full textAbstract:
Tropical coral reefs are among the most diverse ecosystems of the world but facing increasing threats to their health. Over the last thirty years, many Caribbean coral reefs have undergone dramatic changes and experienced large losses in coral cover, due to direct and indirect anthropogenic disturbances. The results of which are reefs with low rugosity, changed trophic dynamics and low fish diversity. In recent times reefs have failed to recover from disturbances due to an increase in frequency and severity of disturbances and stresses. In the Caribbean on many coral reefs this has resulted in a shift towards macroalgal dominance by species of the phylum Phaeophyta. The processes and factors affecting the standing crop of macroalgae are many and complex. Two main hypotheses are identified in the literature as being the driving forces of algal dynamics: nutrient dynamics (availability, supply and uptake) and herbivory. However, many studies have been found to be inconclusive because of the complexity of the coral reef ecosystem, which makes it difficult if not impossible to control for all factors and processes influencing the standing crop of macroalgae such as light, water flow and sedimentation. The inherent characteristics of macroalgae, like morphology and life history, make them behave differently. Whilst herbivore characteristics, like size of mouth parts, feeding modes and preferences, will influence the amount of algal biomass removed. The spatial context (i.e. coral fore reef vs. back reef) will influence the effects of both bottom-up and top-down controls. Besides these inter-habitat differences, macroalgae within similar habitats but differing geographical locations may respond differently, for example, a forereef exposed to the open ocean or a forereef located in a sheltered bay. This thesis attempts to provide insight into the dynamics of two dominant brown macroalgae on Caribbean coral reefs, Dictyota spp. and Lobophora variegata. This aim was addressed by developing a model for the macroalga species Dictyota to model the various processes and factors on a coral forereef affecting percentage cover. Further, the patch dynamics of both Lobophora variegata and Dictyota were investigated to gain an insight into their dynamics under varying environmental conditions: the windward and leeward sides of an atoll. Finally, herbivory is identified as one of the key process affecting macroalgal cover. I investigated this process by deploying cages on both the windward and leeward side of the atoll to investigate the effects of grazing pressure under varying environmental conditions. A Bayesian Belief Network model was developed for Dictyota spp. to model the bottom-up and top-down processes on a coral forereef determining the percentage cover. The model was quantified using relationships identified in the scientific literature and from field data collected over a nine moth period in Belize. This is the first BBN model developed for brown macroalgae. The fully parameterized model identified areas of limited knowledge and because of its probabilistic nature it can explicitly communicate the uncertainties associated with the processes and interactions on standing crop. As such the model may be used as a framework for scientific research or monitoring programmes and it is expected that the model performance to predict macroalgal percentage cover will improve once new information becomes available. Size-based transition matrices were developed for both Dictyota spp. and Lobophora variegata to investigate the patch dynamics under varying environmental conditions: the windward and leeward sides of an atoll. The matrices reveal that standard measures of algal percent cover might provide a misleading insight into the underlying dynamics of the species. Modelling the patch dynamics with matrices provided insight into the temporal behaviour of macroalgae. This is an important process to understand because patch dynamics are determining competitive interactions with other coral reef benthic organisms. The outcome of competitive interactions will differ with macroalgal species. This study indicate that Dictyota spp. responded strongly to differing environmental conditions in that it has reduced growth rates and lower percent cover on the leeward side of the atoll, whilst Lobophora variegata showed far less sensitivity to environmental conditions. The patch dynamics of Dictyota spp. also showed a higher temporal variation than Lobophora variegata but only on the exposed forereef. A caging experiment was set up to investigate the response of both macroalgal species to different grazing pressure scenarios, under varying environmental conditions. Dictyota spp. had a significant response to environmental conditions in that a higher percentage cover was found on the exposed side of the atoll, whilst for Lobophora variegata the response was far less obvious. The less clear response of Lobophora variegata was very likely caused by competition of Dictyota with Lobophora due to the very high cover Dictyota obtained in the cages where all herbivores were excluded. The low grazing pressure treatments also showed an increase in cover of Dictyota, whilst for Lobophora, only a reduction in the rate of increase could be observed. The results indicate that on the leeward side of the atoll, fish grazing alone seems sufficient to control the standing crop of Dictyota and Lobophora variegata. Retrospective analysis of the experimental design showed that the limited size of the experimental set up could have confounded the results for Lobophora as well. In future experiments it is recommended to increase number replicates. Management of coral reef habitats is frequently constrained by a lack of funds and resources. The BBN Model once fully parameterized can provide a useful tool for coral reef management, because the model allows exploration of different reef scenario’s, which in turn can aid in prioritizing management strategies. Furthermore, the thesis provided an insight into the complexities of macroalgal dynamics. The responses of macroalgae to physiological factors and ecological processes are species specific and dependent on the location, and caution against generalizing on what controls the standing crop of macroalgae. Therefore it is argued that future investigations into algal ecology should clearly define the species, habitat and location. This can help to make informed management decisions.
23
Odimabo, Onengiyeofori. "Risk management system to guide building construction projects in developing countries : a case study of Nigeria." Thesis, University of Wolverhampton, 2016. http://hdl.handle.net/2436/618537.
Full textAbstract:
Project risk assessment is an effective tool for planning and controlling cost, time and achieving the technical performance of a building construction project. Construction projects often face a lot of uncertainties, which places building construction projects at the risk of cost, time overruns as well as poor quality delivery. Considering the limited resources of developing countries, there is need to complete building projects on-time, on-budget, and to meet optimal quality hence, risk management is an important part of the decision making process in construction industry as it determines the success or failure of construction projects. In line with this need, this research aims to establish a system to improve the time, cost and quality performance of building construction projects in developing countries, through a comprehensive risk management model that ensures the expectations of clients are met. To achieve the aim of this research, a mixed methodological approach was adopted. Through the review of literature, a conceptual risk management framework suitable to elaborate risk assessment of building construction projects especially for developing countries was developed. A questionnaire survey using a nonprobability sampling technique was conducted to elicit information from construction professionals in Nigeria to assess their perception of 79 risk factors identified from literature review based on the likelihood of occurrence and impact on projects using a five point scale. Responses from 343 construction professionals were drawn from 305 contractors and subcontractors and 38 clients (private and public) within the Nigerian construction sector. Response data was subjected to descriptive statistics to depict the frequency distribution and central tendency of responses. Subsequently, the risk acceptability matrix (RAM) was adopted to categorise and prioritise risk factors. 27 critical risks that affect building construction projects were identified. A Bayesian Belief Network (BBN) model was developed by structural learning and used to examine the cause and effect relationship amongst the 27 critical risk factors. The developed BBN model was subjected to validation using a multiple case study of two building construction projects in Nigeria. The result showed the interrelation between the 27 risk factors and how they contributed to cost and time overruns as well as quality problems. The critical risks directly affecting the cost of building construction project were: fluctuation of material prices; health and safety issues; bribery and corruption; material wastage; poor site management and supervision; and time overruns. The critical factors identified to directly affect quality were: supply of defective materials; working under harsh conditions; improper construction methods; lack of protective equipment; ineffective time allocation; poor communication between involved stakeholders; and unsuitable leadership style. Time overruns on building construction projects was directly caused by: quality problems; low productivity; improper construction methods; poor communication between involved parties; delayed payments in contracts; and poor site management and supervision. As a consolidation of the findings of this research, a BBN model for identifying risk factors that directly affect time, cost and quality on building construction projects has been developed which has the potential for assisting construction stake holders to manage risks on their projects. In view of the findings, a best practice system for risk management in building construction projects in Nigeria has been developed with an implementation guide to help building construction practitioners to successfully implement risk management on their building construction projects. Suitable risk responses, also in the form of recommendations have been identified. The strategies include actions to be taken to respond to risks based on their perceived significance or acceptability as well as some positive risk responses, such as exploiting, sharing, enhancing and accepting, and other negative risk responses, such as avoidance, mitigation transfer and acceptance.
24
Islam, Muhammad Saiful. "Modelling cost overrun risks in power plant projects." Thesis, Queensland University of Technology, 2019. https://eprints.qut.edu.au/125508/1/Muhammad%20Saiful_Islam_Thesis.pdf.
Full textAbstract:
Cost overruns in power plant projects frequently occur, and this very alarming phenomenon requires proper risk assessment and management in the early phase of power plant project development. A modified fuzzy group decision-making approach (FGDMA) was developed and the critical risks in different phases of thermal power plant project were identified. Further, a novel fuzzy canonical model (FCM) was developed and the complex causal risk-networks were modelled to understand the root causes of cost overruns. The benefits of this research to practitioners are such that it provides greater understanding of the risks involved in power plant projects and sound analytical methods to asses the risks.
25
Suermondt, Henri Jacques. "Explanation in Bayesian belief networks." Full text available online (restricted access), 1992. http://images.lib.monash.edu.au/ts/theses/suermondt.pdf.
Full text
26
Olson, John Thomas. "Hardware/software partitioning utilizing Bayesian belief networks." Diss., The University of Arizona, 2000. http://hdl.handle.net/10150/284156.
Full textAbstract:
In heterogeneous systems design, partitioning of the functional specifications into hardware and software components is an important procedure. Often, a hardware platform is chosen and the software is mapped onto the existing partial solution, or the actual partitioning is performed in an ad hoc manner. The partitioning approach presented here is novel in that it uses Bayesian Belief Networks (BBNs) to categorize functional components into hardware and software classifications. The BBN's ability to propagate evidence permits the effects of a classification decision made about one function to be felt throughout the entire network. In addition, because BBNs have a belief of hypotheses as their core, a quantitative measurement as to the correctness of a partitioning decision is achieved. In this research, the motivation and background material are presented first. Next, a methodology for automatically generating the qualitative, structural portion of BBN, and the quantitative link matrices is given. Lastly, a case study of a programmable thermostat is developed to illustrate the BBN approach. The outcomes of the partitioning process are discussed and placed in a larger design context, called model-based Codesign.
27
Narasimha, Rajesh. "Application of Information Theory and Learning to Network and Biological Tomography." Diss., Georgia Institute of Technology, 2007. http://hdl.handle.net/1853/19889.
Full textAbstract:
Studying the internal characteristics of a network using measurements obtained from endhosts is known as network tomography. The foremost challenge in measurement-based approaches is the large size of a network, where only a subset of measurements can be obtained because of the inaccessibility of the entire network. As the network becomes larger, a question arises as to how rapidly the monitoring resources (number of measurements or number of samples) must grow to obtain a desired monitoring accuracy. Our work studies the scalability of the measurements with respect to the size of the network. We investigate the issues of scalability and performance evaluation in IP networks, specifically focusing on fault and congestion diagnosis. We formulate network monitoring as a machine learning problem using probabilistic graphical models that infer network states using path-based measurements. We consider the theoretical and practical management resources needed to reliably diagnose congested/faulty network elements and provide fundamental limits on the relationships between the number of probe packets, the size of the network, and the ability to accurately diagnose such network elements. We derive lower bounds on the average number of probes per edge using the variational inference technique proposed in the context of graphical models under noisy probe measurements, and then propose an entropy lower (EL) bound by drawing similarities between the coding problem over a binary symmetric channel and the diagnosis problem. Our investigation is supported by simulation results. For the congestion diagnosis case, we propose a solution based on decoding linear error control codes on a binary symmetric channel for various probing experiments. To identify the congested nodes, we construct a graphical model, and infer congestion using the belief propagation algorithm. In the second part of the work, we focus on the development of methods to automatically analyze the information contained in electron tomograms, which is a major challenge since tomograms are extremely noisy. Advances in automated data acquisition in electron tomography have led to an explosion in the amount of data that can be obtained about the spatial architecture of a variety of biologically and medically relevant objects with sizes in the range of 10-1000 nm A fundamental step in the statistical inference of large amounts of data is to segment relevant 3D features in cellular tomograms. Procedures for segmentation must work robustly and rapidly in spite of the low signal-to-noise ratios inherent in biological electron microscopy. This work evaluates various denoising techniques and then extracts relevant features of biological interest in tomograms of HIV-1 in infected human macrophages and Bdellovibrio bacterial tomograms recorded at room and cryogenic temperatures. Our approach represents an important step in automating the efficient extraction of useful information from large datasets in biological tomography and in speeding up the process of reducing gigabyte-sized tomograms to relevant byte-sized data. Next, we investigate automatic techniques for segmentation and quantitative analysis of mitochondria in MNT-1 cells imaged using ion-abrasion scanning electron microscope, and tomograms of Liposomal Doxorubicin formulations (Doxil), an anticancer nanodrug, imaged at cryogenic temperatures. A machine learning approach is formulated that exploits texture features, and joint image block-wise classification and segmentation is performed by histogram matching using a nearest neighbor classifier and chi-squared statistic as a distance measure.
28
Heather, Adele. "Bayesian belief networks using conditional phase-type distributions." Thesis, University of Ulster, 2001. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.369955.
Full text
29
Likhari, Amitoj S. "Computing a maximal clique using Bayesian belief networks." [Gainesville, Fla.] : University of Florida, 2003. http://purl.fcla.edu/fcla/etd/UFE0000735.
Full text
30
Saad, Ali. "Detection of Freezing of Gait in Parkinson's disease." Thesis, Le Havre, 2016. http://www.theses.fr/2016LEHA0029/document.
Full textAbstract:
Le risque de chute provoqué par le phénomène épisodique de ‘Freeze of Gait’ (FoG) est un symptôme commun de la maladie de Parkinson. Cette étude concerne la détection et le diagnostic des épisodes de FoG à l'aide d'un prototype multi-capteurs. La première contribution est l'introduction de nouveaux capteurs (télémètres et goniomètres) dans le dispositif de mesure pour la détection des épisodes de FoG. Nous montrons que l'information supplémentaire obtenue avec ces capteurs améliore les performances de la détection. La seconde contribution met œuvre un algorithme de détection basé sur des réseaux de neurones gaussiens. Les performance de cet algorithme sont discutées et comparées à l'état de l'art. La troisième contribution est développement d'une approche de modélisation probabiliste basée sur les réseaux bayésiens pour diagnostiquer le changement du comportement de marche des patients avant, pendant et après un épisode de FoG. La dernière contribution est l'utilisation de réseaux bayésiens arborescents pour construire un modèle global qui lie plusieurs symptômes de la maladie de Parkinson : les épisodes de FoG, la déformation de l'écriture et de la parole. Pour tester et valider cette étude, des données cliniques ont été obtenues pour des patients atteints de Parkinson. Les performances en détection, classification et diagnostic sont soigneusement étudiées et évaluéesFreezing of Gait (FoG) is an episodic phenomenon that is a common symptom of Parkinson's disease (PD). This research is headed toward implementing a detection, diagnosis and correction system that prevents FoG episodes using a multi-sensor device. This particular study aims to detect/diagnose FoG using different machine learning approaches. In this study we validate the choice of integrating multiple sensors to detect FoG with better performance. Our first level of contribution is introducing new types of sensors for the detection of FoG (telemeter and goniometer). An advantage in our work is that due to the inconsistency of FoG events, the extracted features from all sensors are combined using the Principal Component Analysis technique. The second level of contribution is implementing a new detection algorithm in the field of FoG detection, which is the Gaussian Neural Network algorithm. The third level of contribution is developing a probabilistic modeling approach based on Bayesian Belief Networks that is able to diagnosis the behavioral walking change of patients before, during and after a freezing event. Our final level of contribution is utilizing tree-structured Bayesian Networks to build a global model that links and diagnoses multiple Parkinson's disease symptoms such as FoG, handwriting, and speech. To achieve our goals, clinical data are acquired from patients diagnosed with PD. The acquired data are subjected to effective time and frequency feature extraction then introduced to the different detection/diagnosis approaches. The used detection methods are able to detect 100% of the present appearances of FoG episodes. The classification performances of our approaches are studied thoroughly and the accuracy of all methodologies is considered carefully and evaluated
31
Gaskell, Alexander Paul. "Sensor managememt in mobile robotics using Bayesian belief networks." Thesis, University of Oxford, 1995. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.282200.
Full text
32
Lampis, Mariapia. "Application of Bayesian Belief Networks to system fault diagnostics." Thesis, Loughborough University, 2010. https://dspace.lboro.ac.uk/2134/6864.
Full textAbstract:
Fault diagnostic methods aim to recognize when a fault exists on a system and to identify the failures which have caused it. The fault symptoms are obtained from readings of sensors located on the system. When the observed readings do not match those expected then a fault can exist. Using the detailed information provided by the sensors a list of the failures that are potential causes of the symptoms can be deduced. In the last decades, fault diagnostics has received growing attention due to the complexity of modern systems and the consequent need of more sophisticated techniques to identify failures when they occur. Detecting the causes of a fault quickly and efficiently means reducing the costs associated with the system unavailability and, in certain cases, avoiding the risks of unsafe operating conditions. Bayesian Belief Networks (BBNs) are probabilistic graphical models that were developed for artificial intelligence applications but are now applied in many fields. They are ideal for modelling the causal relations between faults and symptoms used in fault diagnostic processes. The probabilities of events within the BBN can be updated following observations (evidence) about the system state. In this thesis it is investigated how BBNs can be applied to the diagnosis of faults on a system with a model-based approach. Initially Fault Trees (FTs) are constructed to indicate how the component failures can combine to cause unexpected deviations in the variables monitored by the sensors. The FTs are then converted into BBNs and these are combined in one network that represents the system. The posterior probabilities of the component failures give a measure of which components have caused the symptoms observed. The technique is able to handle dynamics in the system introducing dynamic patterns for the sensor readings in the logic structure of the BBNs. The method is applied to two systems: a simple water tank system and a more complex fuel rig system. The results from the two applications are validated using two simulation codes in C++ by which the system faulty states are obtained together with the failures that cause them. The accuracy of the BBN results is evaluated by comparing the actual causes found with the simulation with the potential causes obtained with the diagnostic method.
33
Carriger, John Fletcher Jr. "Bayesian belief networks for decision analysis in environmental management." W&M ScholarWorks, 2009. https://scholarworks.wm.edu/etd/1539791560.
Full textAbstract:
In decision problems that rely on technical or scientific data, values are often not explicitly considered, resulting in suboptimal environmental management decision-making. Yet, valuation is an integral part of the overall environmental management process. An environmental decision-making framework that places valuation at the forefront of the process is advocated. The application of values to environmental decisions should occur at every phase of analysis, not just the final weighing of decisions. Value-focused thinking will be used here to structure the problem and determine what is important. Management tasks, environmental or otherwise, cannot rely solely on objective criteria. Stakeholder input and values, and regulatory guidelines are normally considered along with relevant monitoring and modeling data output. Though formal risk management normally contains many decision tools, a unified procedure should exist to weigh evidence as well as formally integrate opinion and observation. A decision framework should be a helpful tool to bring together lines of evidence and values necessary to make important and costly decisions. If the decision-making consequences are detrimental, others can understand why a decision was made if a rationale is available. The best way to understand how a decision was made is to present the decision process from a value-focused perspective. Understanding the difference between objectives, alternatives, and criteria in a decision problem and placing value on features of interests should improve current informal environmental management decisions immensely. Though the current work will not explicitly evaluate costs and benefits, an approach that uses Bayesian Belief Networks (BBNs) and influence diagrams (IDs) is proposed. From the value-focused decision analysis, IDs will be created to weigh the evidence of the various alternative actions needed to reach items of value. An ID can be constructed once the major objectives, alternatives, and criteria are identified. The ID construction phase arranges the information determined in the decision analysis so that experts and lay people can evaluate what is important in a problem and how decisions and other factors influence it. Constructing an ID would include mapping the causal factors and decisions in a directed acyclic graph while preserving assumptions of conditional independence. The first three chapters of this thesis synthesized information from the decision analysis literature to establish an approach that will be beneficial to environmental management. The final two chapters developed examples of the approach that applies Bayesian decision networks in environmental management. Two topics in the final chapters were used to illustrate the framework's potential effectiveness: pesticide ecological risk assessment and natural resource management of Chesapeake Bay seagrass. The pesticide risk management scheme incorporated risk assessment evidence from various models to balance ecological risk management with spraying efficacy judgments. The seagrass assessment evaluated the ability of a BBN to assimilate water quality monitoring data in decision-making that reflect remedial goals. Assessing outcomes and the influences of future processes on restoration targets can be accomplished within the framework of a formal decision analysis with Bayesian networks.
34
Rababah, Osama. "Quality assessment of e-commerce websites using Bayesian belief networks." Thesis, Loughborough University, 2007. https://dspace.lboro.ac.uk/2134/8011.
Full textAbstract:
This thesis raises the issue of quality in E-commerce websites and addresses methodologies and approaches to standardize their assessment. The thesis blends the knowledge of academic research with the opinions and insights from experts and practitioners in the field to provide a comprehensive view of the issues and their remedies. The most experienced and successful E-commerce companies are beginning to realize that key determinants of success or failure are not merely a web presence or a low price but delivering a high quality website. Recent research shows that price and promotion are no longer the main draws for customers to make a decision on a purchase. More sophisticated online customers would rather pay a higher price to a provider with high quality service. Given that the establishment of an E-commerce website is mainly a software development effort; there are several standards that apply in governing the quality of such development. There seems to be an almost overwhelming abundance of quality standards that lead to a high level of cynicism and skepticism surrounding them and the eventual lack of use. Furthermore, no standard can directly predict the quality a website under development is going to achieve. Past approaches concerning the quality of E-commerce websites emphasize usability standards using techniques like feature inspection and collecting data about end-users' opinion by questionnaires. These methods provide important feedback and their results can be utilized as a useful background for future work, however, they do not contribute directly to a dynamic model that enables forecasting. The study of quality in the domain of the Internet in general, and E-commerce in particular, poses new challenges as technology evolves, including methods and metrics for estimating, managing quality during the product life cycle and quality-of-use measurement. The solution proposed by this research is to use a Bayesian Belief Network model. This model provides a consistent and practical approach to assessing the quality of the website. The assessment can be carried out before the completion of the website development, thus, providing insight on the trend and direction for correction and improvements. It can also be carried out on completed and operational work, providing analysis of areas for improvement. The model should be relatively quick and practical in providing an overall comprehensive assessment with root-cause analysis that would lead to corrective measures to improve the quality of the E-commerce website. In this research idioms were applied in realizing a complete Bayesian Belief Network model. The conclusions are measured against comparative assessment to validate the practical benefits of the work accomplished. The WebQual method was utilized to validate the "belief" established by the model.
35
Beaver, Justin. "A LIFE CYCLE SOFTWARE QUALITY MODEL USING BAYESIAN BELIEF NETWORKS." Doctoral diss., University of Central Florida, 2006. http://digital.library.ucf.edu/cdm/ref/collection/ETD/id/2353.
Full textAbstract:
Software practitioners lack a consistent approach to assessing and predicting quality within their products. This research proposes a software quality model that accounts for the influences of development team skill/experience, process maturity, and problem complexity throughout the software engineering life cycle. The model is structured using Bayesian Belief Networks and, unlike previous efforts, uses widely-accepted software engineering standards and in-use industry techniques to quantify the indicators and measures of software quality. Data from 28 software engineering projects was acquired for this study, and was used for validation and comparison of the presented software quality models. Three Bayesian model structures are explored and the structure with the highest performance in terms of accuracy of fit and predictive validity is reported. In addition, the Bayesian Belief Networks are compared to both Least Squares Regression and Neural Networks in order to identify the technique is best suited to modeling software product quality. The results indicate that Bayesian Belief Networks outperform both Least Squares Regression and Neural Networks in terms of producing modeled software quality variables that fit the distribution of actual software quality values, and in accurately forecasting 25 different indicators of software quality. Between the Bayesian model structures, the simplest structure, which relates software quality variables to their correlated causal factors, was found to be the most effective in modeling software quality. In addition, the results reveal that the collective skill and experience of the development team, over process maturity or problem complexity, has the most significant impact on the quality of software products.Ph.D.
School of Electrical Engineering and Computer Science
Engineering and Computer Science
Computer Engineering
36
HUSSAIN, KAMALY MAHBUB. "Bayesian belief networks for guidedremote diagnostics and troubleshootingof heavy vehicles." Thesis, KTH, Maskinkonstruktion (Inst.), 2017. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-209906.
Full textAbstract:
Kostnadreducering och eektivisering av reparationer (t.ex i bilindustrin) har varit malet for forskningen kring guidad diagnostik i snart tvadecennier [ 1], med en onskan till intuitiv felsokning och reparation utan tidigare expert kunskaper. Detta betyder att automation vid diagnostik har blivit en nodvandighet dar det ar mojligt att forstakomplexa system samtidigt som operatoren ges tillrackligt med stod och expertkunskaper fr att kunna tillfora kompetent assistans. Detta examensarbete som utfordes paScania CV AB undersoker hur ett sadant system skulle utformas och prestera samtidigt som arbetet ligger till grund for vidare utveckling av guidad fjarrdiagnostik hos Scania. Resultatet kommer att behandla tre analysomraden. Ett, dem observationer fran fordonet som ar indikationer om ett felaktigt system. Tva, anvandning av ett Basianskt natverk for att gora en diagnos pasystemet samt undersoka hurvida tillvagagangasattet ar eektivt eller inte for den intiutiva kanslan. Tre, en studie och implementation av en eektiv felsokningsalgoritm som minimerar reparationskostnaden baserad paden givna diagnosen, kostnad for reparationav komponenter samt reparationstiden. Examensarbetet kommer forst att presenteras med en djupgaende teoridel och foljs av implementation av teorin till en funktionell prototyp.Intuitive troubleshooting and fault repair without the need of prior expert knowledge of automobiles has become essential in an aim for cost-minimization and eectiveness of repairs, it has been a focus in troubleshooting research for the past decade or two[ 1]. This calls for an automated diagnosis system that is simple to understand and operate whilst at the same time provides the operator with the expert knowledge required for competent assistance. Thismaster thesis conducted at Scania CV AB will investigate how such a system would function and perform, providing a ground work for further development. The result will incorporate three aspects of analysis. First, the observations from the vehicle indicating that something is wrong or faulty. Second, the use of a Bayesian network, a model structure that describes probabilistic relationships and dependencies among system variables, for diagnostic purposes and to examine its haul on intuitive understanding of the system faults. Third, an implementation and study of a troubleshooting algorithm that will minimize the cost of repair based on an easy calculated metric that takes into consideration the probability of fault, cost of observation and the cost of repair (and indirectly also the mean repair time). Given a particular diagnosis, an optimized action plan and repair sequence is given. A thorough review of the underlying theory will be provided for the reader in the rst part of the report, where a slight deviation will be made to further investigate the use of Bayesian lters and its eect on the a priori probabilities used in the Bayesian model. In the nal part the reader will be guided through the implementation of the given theory and emersion of a working prototype.
37
Ang, Kwang Chien. "Applying Bayesian belief Networks in Sun Tzu's Art of Wa /." Monterey, Calif. : Springfield, Va. : Naval Postgraduate School ; Available from National Technical Information Service, 2004. http://library.nps.navy.mil/uhtbin/hyperion/04Dec%5FAng.pdf.
Full text
38
Siraj, Tammeen. "Seismic risk assessment of high-voltage transformers using Bayesian belief networks." Thesis, University of British Columbia, 2013. http://hdl.handle.net/2429/44245.
Full textAbstract:
Past earthquake records showed that a large magnitude earthquake can cause severe damage to high-voltage substations, which may lead to power disruption for a significant amount of time. A high-voltage transformer is one of the key components of a substation. This thesis proposes a probabilistic framework using Bayesian belief network (BBN) model to predict the vulnerability of a high-voltage transformer for a seismic event. BBN has many capabilities that make it well suited for the proposed risk assessment method. This thesis considers past studies, expert knowledge and reported causes of failures to develop an initial integrated risk assessment framework that acknowledges multiple failure modes. Therefore, the framework incorporates major causes of transformer vulnerability due to seismicity, such as liquefaction, rocking response of transformer, or interaction between interconnected equipment. To demonstrate the application of this framework, this thesis elaborates each step of the framework. Finally, the sensitivity analysis was carried out to evaluate the effects of input variables on transformer damage. The paper also illustrates two predictive models using response surface method (RSM) and Markov chain. The proposed framework is particularly handy to perform, and the results can be useful to support decisions on mitigation measures and seismic risk prediction.
39
Luo, Zhiyuan. "A probabilistic reasoning and learning system based on Bayesian belief networks." Thesis, Heriot-Watt University, 1992. http://hdl.handle.net/10399/1490.
Full text
40
Masood, Adnan. "Measuring Interestingness in Outliers with Explanation Facility using Belief Networks." NSUWorks, 2014. http://nsuworks.nova.edu/gscis_etd/232.
Full textAbstract:
This research explores the potential of improving the explainability of outliers using Bayesian Belief Networks as background knowledge. Outliers are deviations from the usual trends of data. Mining outliers may help discover potential anomalies and fraudulent activities. Meaningful outliers can be retrieved and analyzed by using domain knowledge. Domain knowledge (or background knowledge) is represented using probabilistic graphical models such as Bayesian belief networks. Bayesian networks are graph-based representation used to model and encode mutual relationships between entities. Due to their probabilistic graphical nature, Belief Networks are an ideal way to capture the sensitivity, causal inference, uncertainty and background knowledge in real world data sets. Bayesian Networks effectively present the causal relationships between different entities (nodes) using conditional probability. This probabilistic relationship shows the degree of belief between entities. A quantitative measure which computes changes in this degree of belief acts as a sensitivity measure .
The first contribution of this research is enhancing the performance for measurement of sensitivity based on earlier research work, the Interestingness Filtering Engine Miner algorithm. The algorithm developed (IBOX - Interestingness based Bayesian outlier eXplainer) provides progressive improvement in the performance and sensitivity scoring of earlier works. Earlier approaches compute sensitivity by measuring divergence among conditional probability of training and test data, while using only couple of probabilistic interestingness measures such as Mutual information and Support to calculate belief sensitivity. With ingrained support from the literature as well as quantitative evidence, IBOX provides a framework to use multiple interestingness measures resulting in better performance and improved sensitivity analysis. The results provide improved performance, and therefore explainability of rare class entities. This research quantitatively validated probabilistic interestingness measures as an effective sensitivity analysis technique in rare class mining. This results in a novel, original, and progressive research contribution to the areas of probabilistic graphical models and outlier analysis.
41
Goodson, Justin. "Assessing the quality of care in nursing homes through Bayesian belief networks." Diss., Columbia, Mo. : University of Missouri-Columbia, 2005. http://hdl.handle.net/10355/4286.
Full textAbstract:
Thesis (M.S.)--University of Missouri-Columbia, 2005.The entire dissertation/thesis text is included in the research.pdf file; the official abstract appears in the short.pdf file (which also appears in the research.pdf); a non-technical general description, or public abstract, appears in the public.pdf file. Title from title screen of research.pdf file viewed on (July 13, 2006) Includes bibliographical references.
42
Chong, H. G. "Predicting and diagnosing faults in wastewater treatment process by Bayesian Belief Networks." Thesis, Aston University, 1997. http://publications.aston.ac.uk/14154/.
Full textAbstract:
Diagnosing faults in wastewater treatment, like diagnosis of most problems, requires bi-directional plausible reasoning. This means that both predictive (from causes to symptoms) and diagnostic (from symptoms to causes) inferences have to be made, depending on the evidence available, in reasoning for the final diagnosis. The use of computer technology for the purpose of diagnosing faults in the wastewater process has been explored, and a rule-based expert system was initiated. It was found that such an approach has serious limitations in its ability to reason bi-directionally, which makes it unsuitable for diagnosing tasks under the conditions of uncertainty. The probabilistic approach known as Bayesian Belief Networks (BBNS) was then critically reviewed, and was found to be well-suited for diagnosis under uncertainty. The theory and application of BBNs are outlined. A full-scale BBN for the diagnosis of faults in a wastewater treatment plant based on the activated sludge system has been developed in this research. Results from the BBN show good agreement with the predictions of wastewater experts. It can be concluded that the BBNs are far superior to rule-based systems based on certainty factors in their ability to diagnose faults and predict systems in complex operating systems having inherently uncertain behaviour.
43
Durá-Bernal, Salvador. "A cortical model of object perception based on Bayesian networks and belief propagation." Thesis, University of Plymouth, 2011. http://hdl.handle.net/10026.1/540.
Full textAbstract:
Evidence suggests that high-level feedback plays an important role in visual perception by shaping the response in lower cortical levels (Sillito et al. 2006, Angelucci and Bullier 2003, Bullier 2001, Harrison et al. 2007). A notable example of this is reflected by the retinotopic activation of V1 and V2 neurons in response to illusory contours, such as Kanizsa figures, which has been reported in numerous studies (Maertens et al. 2008, Seghier and Vuilleumier 2006, Halgren et al. 2003, Lee 2003, Lee and Nguyen 2001). The illusory contour activity emerges first in lateral occipital cortex (LOC), then in V2 and finally in V1, strongly suggesting that the response is driven by feedback connections. Generative models and Bayesian belief propagation have been suggested to provide a theoretical framework that can account for feedback connectivity, explain psychophysical and physiological results, and map well onto the hierarchical distributed cortical connectivity (Friston and Kiebel 2009, Dayan et al. 1995, Knill and Richards 1996, Geisler and Kersten 2002, Yuille and Kersten 2006, Deneve 2008a, George and Hawkins 2009, Lee and Mumford 2003, Rao 2006, Litvak and Ullman 2009, Steimer et al. 2009). The present study explores the role of feedback in object perception, taking as a starting point the HMAX model, a biologically inspired hierarchical model of object recognition (Riesenhuber and Poggio 1999, Serre et al. 2007b), and extending it to include feedback connectivity. A Bayesian network that captures the structure and properties of the HMAX model is developed, replacing the classical deterministic view with a probabilistic interpretation. The proposed model approximates the selectivity and invariance operations of the HMAX model using the belief propagation algorithm. Hence, the model not only achieves successful feedforward recognition invariant to position and size, but is also able to reproduce modulatory effects of higher-level feedback, such as illusory contour completion, attention and mental imagery. Overall, the model provides a biophysiologically plausible interpretation, based on state-of-theart probabilistic approaches and supported by current experimental evidence, of the interaction between top-down global feedback and bottom-up local evidence in the context of hierarchical object perception.
44
Das, Bibhash. "Technical Due Diligence Assessment and Bayesian Belief Networks Methodology for Wind Power Projects." Thesis, Uppsala universitet, Institutionen för geovetenskaper, 2013. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-224063.
Full textAbstract:
A Technical Due Diligence (TDD) investigation is an important step in the process of obtaining financing, or in mergers and acquisitions, for a wind power project. The investigation, the scope of which varies depending on the stage and nature of the project, involves reviewing important documentation relating to different aspects of the project, assessing potential risks in terms of the quality of the information available and suggesting mitigation or other risk management measures where required. A TDD assessment can greatly benefit from increased objectivity in terms of the reviewed aspects as it enables a sharper focus on the important risk elements and also provides a better appreciation of the investigated parameters. This master’s thesis has been an attempt to introduce more objectivity in the technical due diligence process followed at the host company. Thereafter, a points-based scoring system was devised to quantify the answered questions. The different aspects under investigation have a complex interrelationship and the resulting risks can be viewed as an outcome of a causal framework. To identify this causal framework the concept of Bayesian Belief Networks has been assessed. The resulting Bayesian Networks can be considered to provide a holistic framework for risk analysis within the TDD assessment process. The importance of accurate analysis of likelihood information for accurate analysis of Bayesian analysis has been identified. The statistical data set for the right framework needs to be generated to have the right correct setting for Bayesian analysis in the future studies. The objectiveness of the TDD process can be further enhanced by taking into consideration the capability of the investing body to handle the identified risks and also benchmarking risky aspects with industry standards or historical precedence.
45
Kevorkian, Christopher George. "UAS Risk Analysis using Bayesian Belief Networks: An Application to the VirginiaTech ESPAARO." Thesis, Virginia Tech, 2016. http://hdl.handle.net/10919/73047.
Full textAbstract:
Small Unmanned Aerial Vehicles (SUAVs) are rapidly being adopted in the National Airspace (NAS) but experience a much higher failure rate than traditional aircraft. These SUAVs are quickly becoming complex enough to investigate alternative methods of failure analysis. This thesis proposes a method of expanding on the Fault Tree Analysis (FTA) method to a Bayesian Belief Network (BBN) model. FTA is demonstrated to be a special case of BBN and BBN can allow for more complex interactions between nodes than is allowed by FTA. A model can be investigated to determine the components to which failure is most sensitive and allow for redundancies or mitigations against those failures. The introduced method is then applied to the Virginia Tech ESPAARO SUAV.Master of Science
46
Selvatici, Antonio Henrique Pinto. "Construção de mapas de objetos para navegação de robôs." Universidade de São Paulo, 2009. http://www.teses.usp.br/teses/disponiveis/3/3141/tde-01072009-153749/.
Full textAbstract:
Como a complexidade das tarefas realizadas por robôs móveis vêm aumentando a cada dia, a percepção do robô deve ser capaz de capturar informações mais ricas do ambiente, que permitam a tomada de decisões complexas. Entre os possíveis tipos de informação que podem ser obtidos do ambiente, as informações geométricas e semânticas têm papéis importantes na maioria das tarefas designadas a robôs. Enquanto as informações geométricas revelam como os objetos e obstáculos estão distribuídos no espaço, as informações semânticas capturam a presença de estruturas complexas e eventos em andamento no ambiente, e os condensam em descrições abstratas. Esta tese propõe uma nova técnica probabilística para construir uma representação do ambiente baseada em objetos a partir de imagens capturadas por um robô navegando com uma câmera de vídeo solidária a ele. Esta representação, que fornece descrições geométricas e semânticas de objetos, é chamada O-Map, e é a primeira do gênero no contexto de navegação de robôs. A técnica de mapeamento proposta é também nova, e resolve concomitantemente os problemas de localização, mapeamento e classificação de objetos, que surgem quando da construção de O-Maps usando imagens processadas por detectores imperfeitos de objetos e sem um sensor de localização global. Por este motivo, a técnica proposta é chamada O-SLAM, e é o primeiro algoritmo que soluciona simultaneamente os problemas de localização e mapeamento usando somente odometria e o resultado de algoritmos de reconhecimento de objetos. Os resultados obtidos através da aplicação de O-SLAM em imagens processadas por uma técnica simples de detecção de objetos mostra que o algoritmo proposto é capaz de construir mapas que descrevem consistentemente os objetos do ambiente, dado que o sistema de visão computacional seja capaz de detectá-los regularmente. Em particular, O-SLAM é eficaz em fechar voltas grandes na trajetória do robô, e obtém sucesso mesmo se o sistema de detecção de objetos posuir falhas, relatando falsos positivos e errando a classe do objeto algumas vezes, consertando estes erros. Dessa forma, O-SLAM é um passo em direção à solução integrada do problema de localização, mapeamento e reconhecimento de objetos, a qual deve prescindir de um sistema pronto de reconhecimento de objetos e gerar O-Maps somente pela fusão de informações geométricas e visuais obtidas pelo robô.As tasks performed by mobile robots are increasing in complexity, robot perception must be able to capture richer information from the environment in order to allow complex decision making. Among the possible types of information that can be retrieved from the environment, geometric and semantic information play important roles in most of the tasks assigned to robots. While geometric information reveals how objects and obstacles are distributed in space, semantic information captures the presence of complex structures and ongoing events from the environment and summarize them in abstract descriptions. This thesis proposes a new probabilistic technique to build an object-based representation of the robot surrounding environment using images captured by an attached video camera. This representation, which provides geometric and semantic descriptions of the objects, is called O-Map, and is the first of its kind in the robot navigation context. The proposed mapping technique is also new, and concurrently solves the localization, mapping and object classification problems arisen from building O-Maps using images processed by imperfect object detectors and no global localization sensor. Thus, the proposed technique is called O-SLAM, and is the first algorithm to solve the simultaneous localization and mapping problem using solely odometers and the output from object recognition algorithms. The results obtained by applying O-SLAM to images processed by simple a object detection technique show that the proposed algorithm is able to build consistent maps describing the objects in the environment, provided that the computer vision system is able to detect them on a regular basis. In particular, O-SLAM is effective in closing large loops in the trajectory, and is able to perform well even if the object detection system makes spurious detections and reports wrong object classes, fixing these errors. Thus, O-SLAM is a step towards the solution of the simultaneous localization, mapping and object recognition problem, which must drop the need for an off-the-shelf object recognition system and generate O-Maps only by fusing geometric and appearance information gathered by the robot.
47
Anderson, Jerone S. "A Study of Nutrient Dynamics in Old Woman Creek Using Artificial Neural Networks and Bayesian Belief Networks." Ohio University / OhioLINK, 2009. http://rave.ohiolink.edu/etdc/view?acc_num=ohiou1242921000.
Full text
48
Agrawal, Prerna. "Tool And Algorithms for Rapid Source Term Prediction (RASTEP) Based on Bayesian Belief Networks." Thesis, Uppsala universitet, Institutionen för informationsteknologi, 2015. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-256964.
Full textAbstract:
In case of an accident in a nuclear power plant (NPP), the fast and cor-rect identification of the NPP state that would give a prediction of a possible radioactive release presents a major challenge to both nuclear power plants and regulators. Such prediction is important so that correct and timely decisions and measures are taken to mitigate accident consequences, such as evacuation of people from areas around the power plant. Recent research work [2][3] proposes analyzing the NPP using the Bayesian Belief Network models as a solution to this problem. A BBN is a graphical model that represents any entity with a set of connected nodes. These nodes represent the random variables and the connections between the nodes represent the conditional dependencies between them [7]. However, the BBN models alone are not suitable for use in off-site locations under high stress conditions by people who are not experts. Hence there arises a need for an interface that would – - Be easy to operate by non-experts under high stress situations with incomplete knowledge of the plant state. - Provide the more detailed information about the network that is not easy for users to read out from the BBN itself. - Provide good graphical displays of the radioactive release predictions and other statistics of the network. One such tool is developed as a part of this master thesis project. The contribution is twofold – - Analyzing the user requirements, designing the architecture and development of the tool. - Design and implementation of the algorithms for extracting additional information from the network which is not easy to read out while working directly with the BBN. This kind of information helps the user to take some decisions with entering the observations when the user is not a BBN expert. For instance, it helps the user to know which nodes are important to answer and which nodes can be left out. This also helps the user to interpret the intermediate state of the BBN model of the plant. The tool and the algorithms were evaluated by an expert user in order to assess them based on ease of use, value of the analysis output and the processing time. This project work was carried forward in collaboration with Swedish Radiation Safety Authority (SSM) [8]. SSM is already assessing the tool with the goal to obtain fast and independent predictions of radioactive releases based on plant observations.
49
Zhao, Xi. "3D face analysis : landmarking, expression recognition and beyond." Phd thesis, Ecole Centrale de Lyon, 2010. http://tel.archives-ouvertes.fr/tel-00599660.
Full textAbstract:
This Ph.D thesis work is dedicated to automatic facial analysis in 3D, including facial landmarking and facial expression recognition. Indeed, facial expression plays an important role both in verbal and non verbal communication, and in expressing emotions. Thus, automatic facial expression recognition has various purposes and applications and particularly is at the heart of "intelligent" human-centered human/computer(robot) interfaces. Meanwhile, automatic landmarking provides aprior knowledge on location of face landmarks, which is required by many face analysis methods such as face segmentation and feature extraction used for instance for expression recognition. The purpose of this thesis is thus to elaborate 3D landmarking and facial expression recognition approaches for finally proposing an automatic facial activity (facial expression and action unit) recognition solution.In this work, we have proposed a Bayesian Belief Network (BBN) for recognizing facial activities, such as facial expressions and facial action units. A StatisticalFacial feAture Model (SFAM) has also been designed to first automatically locateface landmarks so that a fully automatic facial expression recognition system can be formed by combining the SFAM and the BBN. The key contributions are the followings. First, we have proposed to build a morphable partial face model, named SFAM, based on Principle Component Analysis. This model allows to learn boththe global variations in face landmark configuration and the local ones in terms of texture and local geometry around each landmark. Various partial face instances can be generated from SFAM by varying model parameters. Secondly, we have developed a landmarking algorithm based on the minimization an objective function describing the correlation between model instances and query faces. Thirdly, we have designed a Bayesian Belief Network with a structure describing the casual relationships among subjects, expressions and facial features. Facial expression oraction units are modelled as the states of the expression node and are recognized by identifying the maximum of beliefs of all states. We have also proposed a novel method for BBN parameter inference using a statistical feature model that can beconsidered as an extension of SFAM. Finally, in order to enrich information usedfor 3D face analysis, and particularly 3D facial expression recognition, we have also elaborated a 3D face feature, named SGAND, to characterize the geometry property of a point on 3D face mesh using its surrounding points.The effectiveness of all these methods has been evaluated on FRGC, BU3DFEand Bosphorus datasets for facial landmarking as well as BU3DFE and Bosphorus datasets for facial activity (expression and action unit) recognition.
50
Punyamurthula, Sudhir. "BAYESIAN-INTEGRATED SYSTEM DYNAMICS MODELLING FOR PRODUCTION LINE RISK ASSESSMENT." UKnowledge, 2018. https://uknowledge.uky.edu/me_etds/124.
Full textAbstract:
Companies, across the globe are concerned with risks that impair their ability to produce quality products at a low cost and deliver them to customers on time. Risk assessment, comprising of both external and internal elements, prepares companies to identify and manage the risks affecting them. Although both external/supply chain and internal/production line risk assessments are necessary, internal risk assessment is often ignored. Internal risk assessment helps companies recognize vulnerable sections of production operations and provide opportunities for risk mitigation.
In this research, a novel production line risk assessment methodology is proposed. Traditional simulation techniques fail to capture the complex relationship amongst risk events and the dynamic interaction between risks affecting a production line. Bayesian- integrated System Dynamics modelling can help resolve this limitation. Bayesian Belief Networks (BBN) effectively capture risk relationships and their likelihoods. Integrating BBN with System Dynamics (SD) for modelling production lines help capture the impact of risk events on a production line as well as the dynamic interaction between those risks and production line variables. The proposed methodology is applied to an industrial case study for validation and to discern research and practical implications.